Enter the text here that is the new abstract information for your application. This section must be no longer than 30 lines of text. Mitotic crossing over is detrimental because it can lead to loss of heterozygosity or chromosome rearrangement, both of which are associated with cancer. The dangers of mitotic crossing over are evident in persons with Bloom syndrome, a rare, hereditary disease characterized by a highly elevated incidence and early onset of a broad range of malignancies. A predominant feature of cells from these patients is elevated mitotic crossing over between sister chromatids, homologous chromosomes, and heterologous chromosomes. Studies of BLM, which encodes a RecQ helicase, are providing important insights into mechanisms cells use to prevent mitotic crossing over. We have made significant contributions to these insights by exploiting unique advantages of Drosophila as a model metazoan for genetic and molecular studies of the cellular functions of the BLM helicase. We have developed a detailed model for the function of Drosophila DmBLM in double- strand break repair; we will now test predictions of this model, as well as models proposed by others, using established and novel molecular genetic assays. This includes novel assays designed for use in cultured human cells. Although repair of double-strand breaks may lead to crossing over, most spontaneous mitotic crossovers are thought to arise from problems at the replication fork, including blocks to fork progression, broken forks, and some fork convergences. Based on our studies of lethal phenotypes that occur when both DmBLM and any of three structure-specific DNA endonuclease is absent, we have modified models to explain roles of DmBLM in replication fork repair; we will use innovative methods to test key predictions of these models. We will also combine in vivo and in vitro studies to uncover functions of these three endonucleases (MUS81-MMS4, GEN, and MUS312-SLX1), each of which has been implicated in resolution of Holliday junction intermediates. Importantly, we find that GEN, which appears to play only a secondary role in budding yeast and is absent from fission yeast, has several important functions in Drosophila. Our studies will therefore lead to a greater understanding of the cellular functions of this enzyme and the other putative resolvases. Finally, we will identify additional genes/proteins involved in preventing or promoting mitotic crossovers through studies of candidate genes and through physical interaction experiments. The results from our proposed studies will enhance our understanding of mechanisms through which mitotic crossovers are generated and how cells employ the tumor suppressor protein BLM and other proteins to prevent mitotic crossing over, and how putative Holliday junction resolvases function to promote crossing over.